Memory research in Alzheimer's disease (AD) has typically focused on word-based or verbal stimuli. However, pictures may have more clinical and ecological importance, as patients are commonly faced with visual stimuli, such as people, landmarks, and medications. Research has shown that pictures can help healthy older adults improve discrimination, reduce false memories, and shift to a more conservative response bias. Response bias is one's tendency to respond "old" or "new" on a recognition memory test. AD patients exhibit an abnormally high rate of false memories and a liberal response bias. Preliminary studies suggest that AD patients may also be able to use the distinctive information provided by pictures to improve their discrimination and reduce their number of false memories. The current application will use techniques of experimental psychology and cognitive neuroscience to examine the effect of pictures on memory performance in AD patients. Two behavioral and two event-related potential (ERP) experiments are proposed to evaluate three specific aims. Aim 1 will compare discrimination and response bias for pictures versus words in a single experiment systematically varying pictures and words at study and test. Aim 2 will use standard neuropsychological measures of medial temporal lobe and frontal lobe function to evaluate neuroanatomical regions that may contribute to the effect of pictures on discrimination and response bias in AD. Finally, Aim 3 will use grayscale versus color pictures and a level of processing manipulation to evaluate two hypotheses regarding how pictures might affect memory processes in AD. [unreadable] [unreadable] [unreadable]